Why did my PIC die?

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

Hi,

please help me understanding why my PIC died?

schematic of the reset button : (howto display this on the forum btw?)

http://users.[spam].be/chs/Werk/reset1.jpg

I used low ohmic pull-up in order to protect the PIC from ESD. The R-C network for the same reasons.

Vpic = 3V:

http://users.[spam].be/chs/Werk/3V.jpg

schematic of the PIC :

http://users.[spam].be/chs/Werk/pic1.jpg

I'm using CCS-ICDU40 (usb) programmer. On programming 13V is put on the reset line. Why does my PIC die?

rberek · Joined: 10 Jan 2005 Posts: 207 Location: Ottawa, Canada

Your pull up should be 47K on the reset if you are programming with the ICD.

I this the small resisitor let the reset pin sink way too much current.

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

How is that?

Don't you see the 390K ohms in series with the MCLR and the button.

Ttelmah · Guest

First comment. You show connections being bought out to a header, as though you are going to use LVP ICSP. If so, the PGM line, should have a resistor to GND, otherwise during bootup, when the pin is programmed as an input, the line can float enough to trigger programming.
Second comment, there are a lot of lines going 'off circuit' (SCAN, Dn etc.). What do these connect to, how are these protected?.
I'd seriously consider having a zener across the piezo element, and a resistor to limit any back emf into the PIC. Remember a piezo element, can both convert voltage to sound, and also operate the other way. This is how piezo lighters etc., work. If an impact takes place on the piezo module, it is very possible that this could have damaged the pin to which it is connected, or the rest of the PIC...

Best Wishes

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

The problem is round the MCLR line. When I connected my programmer, my pic dies!

the data and scan lines are protected at the place where ESD sparks can enter the circuit (not shown).

Don't think the buzzer killed the PIC. The cause is around MCLR line, I don't understand how.

Pin RB3 is used for debugging using the CCS C compiler/programmer

rberek · Joined: 10 Jan 2005 Posts: 207 Location: Ottawa, Canada

I apologize. I went only by your verbal description.

asmallri · Joined: 12 Aug 2004 Posts: 1660 Location: Perth, Australia

You have not used standard reset circuitry.

You should not have the 2.2uF capacitor on the reset line in this configuration. It will result in the 12 volt remaining applied to the reset pic after you have programmed the chip which in turn can result in the PIC being erased or erroneously programmed.

390K is too high a value, especially if there is any leakage current with the capacitor.

The Piezo could damage the port pin (not likely).

If R4/R5 is meant to allow you to measure the battery voltage then this is way outside the supported input impedance to the A/D - it should be less than 2.5K - you should consider putting an opamp configured as a voltage follower between the divider and the PIC. If you stay with just a voltage divider (can't see how you can) then you should also add a diode (preferably schottky) from the resistor junction to the PIC VCC rail.

The same goes for Vadap on AN1.
_________________
Regards, Andrew

http://www.brushelectronics.com/software
Home of Ethernet, SD card and Encrypted Serial Bootloaders for PICs!!

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

Thanks for input, I understand your remarks, but I still don't understand why the PIC died.

Humberto · Posted: Thu Jun 15, 2006 6:09 am

rberek you are right.

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

Humberto, for SURE it works following the CCS ICD specs.

I want to know why it does not work how I wired it.

My PIC died because of something and I want to know why.

edit: That Pic still programs but on verify it reads all 0000h on all adresses. So it's dead.

Ttelmah · Guest

I'd say Tvhhr.
The data sheet requires the voltage on Mclr, to rise from Vdd to Vihh in less than 1uSec. I presume that during the transition, the charge pump does not properly start, and may well overheat. Having a capacitor directly on the Mclr pin, will prevent this from happening.

Best Wishes

Humberto · Posted: Thu Jun 15, 2006 7:00 am

Christophe, the technical information available is the one that provides the manufacturer through the data sheets. For this of kind of problems, the help is limited because the complete internal circuit is not available. Anyway I do not understand what is your final interest in knowing why a component fails while it´s working outside specification.
I believe that you would have to direct this question directly to the manufacturer.

Humberto

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

Christophe · Joined: 10 May 2005 Posts: 323 Location: Belgium

ICD messages:

http://users.[spam].be/chs/Werk/pg1.jpg

http://users.[spam].be/chs/Werk/pg2.jpg

asmallri · Joined: 12 Aug 2004 Posts: 1660 Location: Perth, Australia

I am not familar with your programmer..

Are you doing a bulk erase?

The 2.2uF could have resulted in the fuses being set to some value other than what you are expecting. For example the protect fuses could be set.
_________________
Regards, Andrew

http://www.brushelectronics.com/software
Home of Ethernet, SD card and Encrypted Serial Bootloaders for PICs!!